O R I G I N A L P A P E R
Structural analysis of Cu(II) ligation to the 5¢-GMP nucleotide
by pulse EPR spectroscopy
Maria Grazia SantangeloÆ Alfredo Medina-Molner Æ Arthur SchweigerÆ George Mitrikas Æ
Bernhard Spingler
Received: 22 December 2006 / Accepted: 12 March 2007 / Published online: 6 April 2007
SBIC 2007
Abstract Simple copper salts are known to denature poly d(GC). On the other hand, copper complexes of substituted 1,4,7,10,13-pentaazacyclohexadecane-14,16-dione are able to convert the right-handed B form of the same DNA se-quence to the corresponding left-handed Z form. A re-search program was started in order to understand why Cu(II) as an aquated ion melts DNA and induces the conformational change to Z-DNA in the form of an aza-macrocyclic complex. In this paper, we present a contin-uous wave and pulse electron paramagnetic resonance study of the mononucleotide model system Cu(II)–guano-sine 5¢-monophosphate . Pulse EPR methods like electron– nuclear double resonance and hyperfine sublevel correla-tion spectroscopy provide unique informacorrela-tion about the electronic and geometric structure of this model system through an elaborate mapping of the hyperfine and nuclear quadrupole interactions between the unpaired electron of the Cu(II) ion and the magnetic nuclei of the nucleotide ligand. It was found that the Cu(II) ion is directly bound to
N7 of guanosine 5¢-monophosphate and indirectly bound via a water of hydration to a phosphate group. This set of experiments opens the way to more detailed structural characterization of specifically bound metal ions in a variety of nucleic acids of biological interest, in particular to understand the role of the metal–(poly)nucleotide interaction.
Keywords Copper Electron–nuclear double resonance Hyperfine sublevel correlation spectroscopy Pulse electron paramagnetic resonance spectroscopy Nucleotide Abbreviations
CW Continuous wave
DMSO Dimethyl sulfoxide
ENDOR Electron-nuclear double resonance EPR Electron paramagnetic resonance 5¢-GMP Guanosine 5¢-monophosphate HYSCORE Hyperfine sublevel correlation Triflate Trifluoromethanesulfonate
Introduction
The interaction of metal complexes with oligonucleotides is one of the central topics in bioinorganic chemistry. We have been interested in metal complexes that are able to induce Z-DNA [1]. Copper complexes of substituted 1,4,7,10,13-pentaazacyclohexadecane-14,16-dione are able to convert the right-handed B form of poly d(GC) to the corresponding left-handed Z form [2]. Pre-formed Z-DNA crystals that were soaked with copper salts bound the metal ions exclusively via N7 of guanine [3]. On the other hand, simple copper salts are known to
Arthur Schweiger died on 4 January 2006. M. G. Santangelo A. Schweiger G. Mitrikas Laboratory of Physical Chemistry,
ETH Zurich,
8093 Zurich, Switzerland
A. Medina-Molner B. Spingler (&) Institute of Inorganic Chemistry, University of Zurich,
8057 Zurich, Switzerland e-mail: spingler@aci.unizh.ch G. Mitrikas (&)
Institute of Materials Science,
NCSR ‘‘Demokritos’’, Aghia Paraskevi Attikis, 15310 Athens, Greece
e-mail: george.mitrikas@phys.chem.ethz.ch DOI 10.1007/s00775-007-0230-1
denature the same DNA sequence [4]. Sundaralingam and Carrabine [5] proposed a model to explain this behavior. We started a research program in order to understand why copper(II) as an aquated ion denatures DNA on the one hand and induces the conformational change to Z-DNA in the form of an azamacrocyclic complex on the other hand. Copper is an ideal metal for electron paramagnetic resonance (EPR) spectroscopy [6–8]. In order to successfully study the interaction between copper and oligonucleotides, one has first to study the copper– mononucleotide system. We were rather surprised that this has not yet been done with the help of pulse EPR [9– 11]. These techniques allow, together with the use of isotope-labeled nucleotides, the determination of the dis-tance between the paramagnetic copper center and se-lected atoms around it within a radius of not more than 6 A˚ . In this paper, based on pulse EPR experiments, we are able to describe the geometry of copper(II) bound to guanosine 5¢-monophosphate (5¢-GMP) in solution. This study is the entry point to further characterize copper– oligonucleotide systems.
Materials and methods Sample preparation
We prepared four different samples: sample 1, copper(II) trifluoromethanesulfonate (triflate) in 50% dimethyl sulf-oxide (DMSO) and 50% H2O; sample 2, copper(II) triflate
and 5¢-GMP (Fig.1a) in 50% DMSO and 50% H2O;
sample 3, copper(II) triflate and 5¢-GMP with deuterium atoms instead of the position 8 (D8-5¢-GMP, Fig.1b) in 50% DMSO and 50% H2O; sample 4, copper(II) triflate
and 5¢-GMP deuterated at all the exchangeable protons (H8-d5-5¢-GMP, Fig.1c) in 50% DMSO-d6and 50% D2O.
All reagents and nucleotides were purchased from Fluka at the highest purity available. In all samples the metal con-centration was 2 mM. For the 5¢-GMP-containing samples (samples 2, 3, and 4) the metal-to-ligand ratio was 1:1. D8-5¢-GMP was prepared by overnight treatment of D8-5¢-GMP at 60C with 4 equiv of triethylamine in D2O. Excess reagent
was removed by repeated lyophilization to dryness from H2O [12]. 1H NMR showed the absence of any
triethyl-amine. The H8-d5-5¢-GMP ligand was prepared by
over-night treatment in D2O. Excess reagent was removed by
lyophilization. The pH of all the samples was determined to be 6.4.
Spectroscopy
Continuous wave (CW) EPR measurements at X-band were carried out using a Bruker E500 spectrometer equipped with a Bruker super-high-Q cavity. Experimental conditions were as follows: microwave frequency, 9.5 GHz; microwave power incident to the cavity, 20 mW; modulation frequency, 100 kHz; modulation amplitude, 0.5 mT. Cooling of the sample was performed with a liquid-nitrogen finger Dewar vessel to 130 K.
The X-band pulse EPR experiments were performed with a Bruker E580 spectrometer (microwave frequency, 9.72 GHz) equipped with a liquid-helium cryostat from Oxford Instruments. All experiments were done at 20 K and a repetition rate of 1 kHz. The magnetic field was measured with a Bruker ER083 CS Gauss meter.
Hyperfine sublevel correlation (HYSCORE) spectros-copy [13,14], with the pulse sequence p/2–s–p/2–t1–p–t2–
p/2–s–echo, is a two-dimensional experiment that corre-lates nuclear frequencies in one electron spin manifold to nuclear frequencies in the other electron spin manifold. The HYSCORE spectra were recorded with the following instrumental parameters: tp/2= tp= 16 ns; starting values
of the two variable times t1and t2, 56 ns; time increment,
Dt = 16 ns (data matrix 350 · 350). To avoid blind spots, spectra with different s values were recorded and added. An eight-step phase cycle was used to remove unwanted echoes.
The Davies electron-nuclear double resonance (EN-DOR) [13,15] experiments were carried out with the pulse sequence p–T–p/2–s–p–s–echo, with microwave pulse lengths of 32, 16, and 32 ns, respectively, and an interpulse time s of 400 ns. A radiofrequency p pulse of variable frequency and a length of 9 ls was applied during time T (10 ls).
The Mims ENDOR [13, 16] experiments were per-formed using the pulse sequence p/2–s–p/2–T–p/2–s–echo, with microwave pulse lengths of 16 ns and an interpulse time of s = 500 ns. During time T (10 ls) a radiofrequency
(a) (b) (c)
Fig. 1 Chemical structures of guanosine 5¢-monophosphate (5¢-GMP) and selectively deuterated 5¢-GMP: a 5¢-GMP; bD8-5¢-GMP; c H8-d5-5¢-GMP
pulse with variable frequency and a length of 9 ls was applied.
Both ENDOR and HYSCORE experiments were carried out at different observer positions that correspond to dif-ferent orientations of the molecules with respect to B0
(orientation selectivity) [13]. Data manipulation
The data were processed with the program MATLAB 7.2 (MathWorks, Natick, MA, USA). The time traces of the HYSCORE spectra were baseline-corrected with a two-order exponential, apodized with a Gaussian window and zero-filled. After a two-dimensional Fourier transformation the absolute-value spectra were calculated. The HYSCORE spectra recorded with different s values were added to eliminate s-dependent blind spots.
The CW EPR and ENDOR spectra were simulated with the EasySpin package [17]. For the CW EPR simulations the hyperfine coupling shift due to the two copper isotopes
63Cu and 65Cu was taken into account. The HYSCORE
spectra were simulated with a program written in-house [18].
Theory
The spin Hamiltonian for a spin system with a Cu(II) ion (electronic configuration 3d9, S = 1/2) and m nuclei with spins I is given by Eq. 1 [13]:
H¼ beB0gS=hþ Xm k SAkIk bn Xm k gn;kB0Ik=h þ X Ik[1=2 IkQkIk; ð1Þ
where the terms describe the electron Zeeman interaction, the hyperfine interaction, the nuclear Zeeman interaction, and the nuclear quadrupole interaction (for nuclei with I > 1/2). The electron Zeeman interaction is characterized by the g tensor that is essentially determined by the metal ion and the directly coordinated ligand atoms. The g values observed in the EPR spectrum together with the metal hyperfine coupling can often be used as a fingerprint to identify the metal ion and to provide information on the symmetry of the paramagnetic center. The ligand hyperfine interaction can be written as the sum of the isotropic interaction or Fermi contact interaction HF= aisoSI and the electron–nuclear dipole–dipole
coupling HDD= STI. Here aiso is the isotropic hyperfine
coupling constant which is directly related to |w0(0)|2, the
electron spin density at the nucleus and matrix T describes the anisotropic dipole–dipole coupling. For protons, the anisotropic part of the hyperfine interaction
can be approximated using the point-dipole model assuming that the distance r between the unpaired electron and the proton is larger than 0.25 nm. In this case, the point-dipole formula (Eq. 2)
T¼ l0
4phgebegnbn
ð3~nn 1Þ
r3 ; ð2Þ
where n is the electron–nucleus unit vector, can be used to calculate the electron–nuclear distance and orientation with respect to the g tensor.
For nuclear spins larger than 1/2 , such those of as14N and2H with I = 1, the nuclear quadrupole principal values [Qx, Qy, Qz] = [–K(1 – g), –K(1 + g), 2K] of the traceless
Qmatrix are usually expressed by the quadrupole coupling constant K = e2qQ/4I(2I – 1)h and the asymmetry param-eter g = (Qx– Qy)/Qz, where Q is the nuclear quadrupole
moment and eq is the electric field gradient.
For an S =1/2, I = 1/2 spin system there are two single-quantum nuclear transitions (|DmS| = 0, |DmI| = 1). For the
special case when g, A, and Q are coaxial and B0is along
one of the principal axes, the first-order frequencies are given by Eq. 3:
mi¼ mj I Ai=2j; ð3Þ
where mIis the nuclear Zeeman frequency and Ai(i = x, y,
z) is one of the principal hyperfine values. For mI> |Ai/2|
(weak coupling case) the two frequencies are centered at mI and separated by Ai. For mI< |Ai/2| (strong coupling
case) the two frequencies are centered at Ai/2 and split by
2mI.
For an S = 1/2, I = 1 spin system there are four single-quantum nuclear frequencies given by Eq. 4,
mi¼ mj I Ai=2 3=2 Qij; ð4Þ
and two double-quantum nuclear transitions (|DmS| = 0,
|DmI| = 2). If the anisotropic hyperfine coupling is small
compared to the isotropic hyperfine interaction and the nuclear quadrupole interaction, the latter frequencies are given by Eq. 5 [19]:
mdqa;b¼ 2 að iso=2 mIÞ2þ K23þ g2
h i1=2
: ð5Þ
ENDOR and HYSCORE experiments are especially suited for accurately measuring nuclear transition fre-quencies which are related to the hyperfine and nuclear quadrupole interactions of the spin system. ENDOR spec-tra consist of peaks that correspond mainly to single-quantum transition frequencies, whereas the HYSCORE spectrum of an S = 1/2, I = 1 system is usually dominated by cross-peaks between the double-quantum frequencies.
Results and discussion
The pKa value of the second deprotonation step of the
phosphate group in 5¢-GMP was determined to be 6.25 ± 0.02 [20]. Addition of 1 equiv of copper lowers this pKavalue down to 4.9 ± 0.3 [21]. The pH of all the
sam-ples measured was 6.4 ± 0.1; therefore, all the Cu(II)–5¢-GMP complexes formed had a neutral charge.
CW EPR spectra
The frozen-solution X-band EPR spectra of the copper(II) triflate sample (sample 1) and the Cu(II)–5¢-GMP samples (samples 2, 3, and 4) together with their simulations are shown in Fig.2. All spectra revealed features that are typical for Cu(II) complexes with aðdx2y2Þ1 ground state (gi> g^) [22] Table1). The comparison of the g values
and the Cu(II) hyperfine parameters between copper(II) triflate (sample 1) and Cu(II)–5¢-GMP (samples 2, 3, and 4) indicates a metal–guanosine interaction. Indeed the EPR parameters of copper(II) triflate are typical for Cu(II) complexes with four oxygen atoms as equatorial ligands [23]. Instead the g values and Cu hyperfine parameters of Cu(II)–5¢-GMP complexes are typical for Cu(II) com-plexes with either three oxygen atoms and one nitrogen atom as equatorial ligands or four oxygen as equatorial ligands. We cannot distinguish between the two possibili-ties by the inspection of CW EPR spectra. However, application of pulse EPR spectroscopy (see ‘‘14N Davies ENDOR spectra’’) ruled out the second possibility, con-firming that the Cu ion is coordinated to three oxygen atoms and one nitrogen atom. This is an indication that the surrounding environment of the Cu(II) ion changed when the 5¢-GMP ligand was added. Moreover the g values and the hyperfine parameters of samples 2, 3, and 4 are very similar to each other. This is an indication that the inter-action between the Cu(II) ion and the guanosine nucleotide is the same for all three slightly different kinds of 5¢-GMP ligands that we used. Although the CW EPR data provide evidence of a coordination of the copper metal by the 5¢-GMP nucleotide, detailed information about the metal ion site cannot be obtained. Further insight into the metal ion site is provided by pulse EPR techniques.
14N Davies ENDOR spectra
The Davies ENDOR spectra of Cu(II)–H8-d5-5¢-GMP
(sample 4) at different magnetic field positions together with their simulations are shown in Fig.3. Davies ENDOR experiments are very well suited for measuring strong hyperfine couplings. The spectra are characterized by two broad peaks centered at approximately |a/2| = 16 MHz and split by twice the nitrogen nuclear Zeeman frequency, 2mN.
This is typical for a strongly coupled nitrogen nucleus with |a/2| > |mN| [13]. This result combined with the CW EPR
parameters validates the hypothesis that we have a copper complex with three oxygen atoms and one nitrogen atom as equatorial ligands. The additional splitting observed in all spectra is assigned to the nuclear quadrupole interaction. Moreover, signals from weakly coupled protons, which overlap with signals from the strongly coupled nitrogen, were suppressed by using short microwave pulses. This effect is known as hyperfine contrast-selective ENDOR [13]. Davies ENDOR measurements at different observer positions (thin traces) and their simulations (thick traces) allow for the estimation of the principal values of the hyperfine, A, and nuclear quadrupole, Q, coupling tensors: (Ax, Ay, Az) = (32.0, 38.0, 31.0) ± 0.2 MHz, and (Qx, Qy,
Qz) = (2.1, –1.5, –0.6) ± 0.5 MHz. Because the larger
hyperfine coupling occurs at the observer position corre-sponding to g^, we conclude that this strongly coupled
nitrogen occupies an equatorial position. Although the N7 atom is the most plausible binding site of the ligand, from these data we are not able to identify, in a definitive way, the identity of this strongly coupled nitrogen. However, with the knowledge of the hyperfine and nuclear quadru-pole interactions between the unpaired electron of the Cu(II) ion and the other magnetic nuclei, it is possible to obtain detailed insight into the local environment of the paramagnetic center, and to validate the hypothesis that the Cu(II) ion is directly coordinated to atom N7.
1H HYSCORE spectra
To get information about the weakly coupled nuclei, the HYSCORE technique was utilized. The X-band proton spectra of samples 2, 3, and 4 are shown in Fig.4together with the corresponding simulations. The HYSCORE spectrum of the Cu(II)–5¢-GMP complex (sample 2, Fig.4a) shows at least two types of weakly coupled pro-tons. One type is characterized by an intense ridge close to the antidiagonal at the proton Larmor frequency (approx-imately 14 MHz) and can be assigned either to protons of water molecules coordinated in axial positions or/and to weakly coupled protons of solvent molecules [24]. The other type of proton HYSCORE signal is distinguished by a broad ridge shifted away from the antidiagonal. This spectrum can be simulated with the hyperfine parameters (Ax, Ay, Az) = (–6.3, –6.3, 9.3) ± 0.2 MHz and Euler
angles [a, b, c] = [0, 90, 0], which correspond to an isotropic hyperfine coupling constant of aiso= –1.1 ±
0.2 MHz and a dipolar coupling constant of T = 5.2 ± 0.2 MHz. In the point-dipole approximation, the latter value corresponds to a point-dipole distance of 2.5 ± 0.1 A˚ . These weakly coupled protons can be assigned either to ligand protons of the nucleotide or to water
molecules directly coordinated to the copper. In order to be able to distinguish between these options we repeated the same experiment for the Cu(II)–D8-5¢-GMP sample (sample 3) with a deuterium nucleus in the position 8 (Fig.1b). New peaks from a deuterium nucleus appeared at low frequency (for details see ‘‘14N, 2H HYSCORE spectra’’; Fig.5b). Moreover, the proton HYSCORE spectrum in Fig.4b can be simulated with the same hyperfine parameters we found for sample 2 (Table2). This finding indicates that in both complexes we can detect the same type of weakly coupled protons. This proton peak certainly cannot be assigned to H8 because in sample 3 there is a deuterium at this position. Moreover the hyper-fine couplings and the distance obtained from our analysis are compatible with the hyperfine couplings and the dis-tance of water protons in Cu[(H2O)6]2+ complexes
deter-mined in a HYSCORE [24] and a single-crystal CW ENDOR study [25]. These considerations confirm the hypothesis that the aforementioned HYSCORE signal can be assigned to protons of water molecules directly coor-dinated to the copper center. To study the hyperfine interaction between the unpaired electron of the copper ion and the H8 proton of the nucleotide, we measured the HYSCORE spectrum of the Cu(II)–H8-d5-5¢-GMP sample
(sample 4) in 50% DMSO-d6and 50% D2O. In this
sam-ple, all the exchangeable protons have been replaced with deuterium nuclei (Fig.1c). The experimental spectrum in Fig.4c consists of proton peaks close to the antidiagonal at the proton Larmor frequency and combination peaks
260 280 300 320 340 B0 / mT (a) (b) (c) (d) A B C D E F
Fig. 2 X-band continuous wave electron paramagnetic resonance spectra taken at 130 K: a Copper(II)trifluoromethanesulfonate in dimethyl sulfoxide (DMSO)/H2O solution (sample 1); b Cu(II)–5¢-GMP complex in DMSO/H2O solution (sample 2); c Cu(II)–D8-5¢-GMP complex in DMSO/H2O solution (sample 3); d Cu(II)-H8-d5 -5¢-GMP complex in DMSO-d6/D2O solution (sample 4). Thin traces experiments, thick traces simulations. A–F observer positions used for the Davies and Mims electron-nuclear double resonance (ENDOR) measurements
Table 1 gvalues and63Cu hyperfine parameters Sample g^(±0.005) gi(±0.005) |A^| (±10) (MHz) |Ai| (±10) (MHz) 1 2.079 2.406 25 382 2 2.072 2.358 15 455 3 2.067 2.340 15 460 4 2.069 2.353 15 455
See ‘‘Sample preparation’’ for an explanation of the samples
5 10 15 20 25 30
νRF/ MHz A
B
C
Fig. 3 X-band Davies ENDOR spectra of Cu(II)–H8-d5-5¢-GMP in DMSO-d6/D2O solution (sample 4) taken at observer positions A–C (see Fig.2, spectra d). Thin traces experiments, thick traces simulations
between the proton and the deuterium frequency. The pro-ton signal can be simulated with the hyperfine parameters (Ax, Ay, Az) = (–0.3, –0.3, 7.5) ± 0.2 MHz and Euler angles
[a, b, c] = [0, 30, 0], which correspond to an isotropic hyperfine coupling constant of aiso = 2.3 ± 0.2 MHz and a
dipolar coupling constant of T = 2.6 ± 0.2 MHz. In the point-dipole approximation, the latter value corresponds to a point-dipole distance of 3.1 ± 0.1 A˚ . This weakly coupled
proton can be exclusively assigned to H8 since in this sample there are no other protons in such close proximity to the copper center. Moreover, our data are in agreement with the imidazole proton measured in a copper–histidine complex by W-band Davies ENDOR [26]. The distance between the H8 proton and the paramagnetic center is compatible with the hypothesis that the Cu(II) ion is directly coordinated to nitrogen N7.
Fig. 4 X-band proton hyperfine sublevel correlation
(HYSCORE) spectra measured at a magnetic field position close to g^. a Cu(II)–5¢-GMP (sample 2), s = 100, 120, 140, and 160 ns; b Cu(II)–D8-5¢-GMP (sample 3), s = 100, 120, 140, and 160 ns; c Cu(II)–H8-d5-5¢-GMP (sample 4); s = 100, 120, 140, 160, 400, 450, and 500 ns; d–f corresponding simulations. The antidiagonal lines are given for m1H
Fig. 5 X-band nitrogen and deuterium HYSCORE spectra measured at a magnetic field position close to g^. a Cu(II)– 5¢-GMP (sample 2), s = 100, 120, 140, and 160 ns; b Cu(II)– D8-5¢-GMP (sample 3), s = 100, 120, 140, and 160 ns; c Cu(II)–H8-d5-5¢-GMP (sample 4); s = 100, 120, 140, 160, 400, 450, and 500 ns; d–f corresponding simulations. The antidiagonal lines are given for m2Hand 2m2H
Table 2 1H hyperfine parameters, Euler angles, and derived distances
Sample A^(±0.2) (MHz) Ai(±0.2) (MHz) [a, b, c] () aiso(±0.2) (MHz) T (±0.2) (MHz) r (±0.1) (A˚ )
2 –6.3 9.3 [0, 90, 0] –1.1 5.2 2.5
3 –6.3 9.3 [0, 90, 0] –1.1 5.2 2.5
14N,2H HYSCORE spectra
The low-frequency regions of the HYSCORE spectra of samples 2, 3, and 4 are shown in Fig.5 together with the corresponding simulations. The HYSCORE spectrum of Cu(II)–5¢-GMP (sample 2) (Fig.5a) is dominated by cross-peaks that are assigned to double-quantum correlation peaks from 14N and their strong intensity is typical for disordered S = 1/2, I = 1 spin systems with negligible anisotropic hyperfine coupling [27]. Numerical simulations of the set of spectra recorded at different observer positions (not shown here) yield the following hyperfine and nuclear quadrupole parameters: (Ax, Ay, Az) = (1.51, 1.51,
1.63) ± 0.05 MHz with Euler angles [a, b, c] = [0, 10, 0] and |e2qQ/h| = 2.44 ± 0.05 MHz with Euler angles [a, b, c] = [90, 90, 0] and g = 0.56 ± 0.05. This nitrogen is characterized by a fairly isotropic hyperfine interaction and a significant quadrupole interaction. The latter can be used as a probe in order to identify this nitrogen. Its comparison with nuclear quadrupole parameters of guanine nitrogens obtained by nuclear quadrupole resonance studies [28] indicates that N1 (e2qQ/h = 2.63 MHz, g = 0.60) is a more plausible candidate than N7 (e2qQ/h = 3.27 MHz, g = 0.16). Although other 5¢-GMP nitrogens, like N9 (e2qQ/h = 1.91 MHz, g = 0.75) or N3 (data not available), cannot be excluded, our nuclear quadrupole parameters together with the weak hyperfine coupling strongly suggest that the signal can be assigned to a remote nitrogen (other than N7) of the 5¢-GMP ligand. This finding is in line with the hypothesis that the Cu(II) ion is directly coordinated to nitrogen N7, as was suggested by the previous analysis of the proton HYSCORE spectra.
The HYSCORE spectrum of the Cu(II)–D8-5¢-GMP sample (sample 3) (Fig.5b) also shows a similar remote nitrogen pattern. This weakly coupled nitrogen is charac-terized by the same hyperfine and nuclear quadrupole parameters of the nitrogen as in sample 2. Moreover, in sample 3, a new cross-peak appears close to the2H Larmor frequency at approximately (2.3, 2.3) MHz, which can be assigned to the deuterium nucleus D8. Since the gyro-magnetic ratio for deuterium is approximately 6.5 times smaller than for a proton, the hyperfine interaction is ex-pected to scale as A(2H) = A(1H)/6.5. By scaling the hyperfine parameters of the H8 proton observed in sam-ple 4, (Ax, Ay, Az) = (–0.05, –0.05, 1.15) MHz and Euler
angles [a, b, c] = [0, 30, 0], and assuming a small quad-rupole interaction with |e2qQ/h| = 0.4 MHz and g = 0.03 (which is typical for deuterium nuclei bonded to carbon), we could accurately simulate the peak assigned to the deuterium nucleus (Fig.5b, e). This cross-check further supports our previous assignment of the proton H8.
In Fig.5c, the HYSCORE spectrum of the Cu(II)–H8-d5-5¢-GMP sample (sample 4) is shown. In this sample we
have abundant deuterium nuclei from the ligand and the solvent. This together with suppression effects often encountered in electron spin echo envelope modulation spectroscopy [29] could account for the absence of the remote nitrogen peaks observed in the other two samples. The spectrum is characterized by single-quantum, double-quantum, and combination peaks of deuterium nuclei. It can be simulated by scaling the hyperfine parameters of the protons detected in samples 2 and 3 (ascribed to water molecules directly coordinated to copper), (Ax, Ay, Az) =
(–0.97, –0.97, 1.43) MHz with Euler angles [a, b, c] = [0, 90, 0], and assuming a negligible quadrupole interaction. This result is in agreement with the assignment of the proton HYSCORE spectra of samples 2 and 3 discussed in the previous section.
31P Mims ENDOR spectra
To detect the interaction between the copper center and the 31P nucleus of the phosphate group, the Mims ENDOR technique was used. This method is very useful to detect weak hyperfine couplings from nuclei that are at relatively long distances from the paramagnetic center (typically r‡ 5 A˚ ) [30]. A weak 31P coupling should
produce a pair of transitions positioned symmetrically about the Larmor frequency for that nucleus. Mims ENDOR spectra of the Cu(II)–5¢-GMP sample (sample 2) recorded at different magnetic field positions and their simulations are shown in Fig. 6. The hyperfine values and Euler angles used for the simulation are (Ax, Ay, Az) =
(–0.20, –0.20, 0.45) ± 0.05 MHz and [a, b, c] = [0, 40, 0]. For this interaction, essentially no electron delocal-ization into the 31P nucleus is seen (Aiso= 0.02 ±
0.02 MHz) and a dipolar coupling of T = 0.22 ± 0.02 MHz is obtained. This dipolar coupling corresponds to a point-dipole distance r = 5.3 ± 0.2 A˚ . The X-ray structure analysis of Cu(II) with 5¢-GMP not only con-tains three Cu–5¢-GMP molecules in the asymmetric unit but also three different types of interactions between the copper ion and the phosphate group [31]. The copper ion binds by direct coordination to the phosphate or via a two hydrogen bridges mediated via two cis-coordinated water molecules, or finally via a single hydrogen bridge to the phosphate mediated by one water molecule that is copper-coordinated. These three different interactions have the following increasing copper to phosphate distance ranges: 3.14–3.27, 4.63, and 5.10–5.52 A˚ . The distance of 5.3 ± 0.2 A˚ that we found by the Mims ENDOR method is clearly compatible with the latter type of metal–phosphorus interaction distances observed for water-mediated phosphate ligation. Therefore, we can conclude that the Cu(II) ion is indirectly bound via a single water molecule to a phosphate group.
Structural considerations
The g values obtained for all the samples indicate that the copper ion in all the samples studied in this work has a ðdx2y2Þ1 ground state with a square-planar symmetry. From Davies ENDOR spectra is not possible to determine the number of nitrogens directly coordinate to the copper center. However, from the CW spectrum (Fig.2, spec-tra d), we can exclude that more than one nitrogen is di-rectly coordinated to the copper because, since in our samples the metal-to-ligand ratio was 1:1, we do not have evidence of the presence of two species that instead would have formed if more nitrogen atoms were directly coordi-nated to the copper center. Combining the results of the Davies ENDOR and CW analyses, we can conclude that the copper ion is directly coordinated to only one nitrogen of the nucleotide. Since this latter nitrogen is characterized by a very large isotropic and a very small anisotropic hyperfine constant, we can say that the nitrogen is equa-torially coordinated to the copper ion. Moreover we can exclude the presence of a dimer because the CW spectra of all the samples did not show any signals at half the mag-netic field (data not shown). A search in the Cambridge Structure Database [32] was undertaken in order to deter-mine to which nitrogen atom(s) of any 9-substituted guanine copper was found to bind. Actually all published X-ray structures of copper binding to a nitrogen of 9-substituted guanine show that the copper is exclusively binding to N7 (e.g., [31,33]). From the proton HYSCORE
spectra we have evidence of a direct coordination of the copper ion by water molecules and in addition by N7 (because of the H8 signal). The molecular model of the Cu(II)–5¢-GMP complex consistent with all our data is shown in Fig. 7.
Conclusions
Metal ion coordination to nucleic acids is essential for the biological function of nucleic acids. In this paper we pre-sented a CW and a pulse EPR study of the mononucleotide model system Cu(II)–5¢-GMP. We obtained a complete characterization of the structural features of the metal ion bound to the nucleotide. The copper is directly coordinated to N7 of the guanine and to five solvent molecules; one of them forms a hydrogen bridge to the phosphate group. This paper provides a basis for the future characterization of copper–oligonucleotide complexes.
Acknowledgements We thank R.K.O. Sigel and C. Finazzo for stimulating discussions. We thank ETH and the Swiss National Sci-ence Foundation for financial support.
References
1. Spingler B (2005) Inorg Chem 44:831–833
2. Spingler B, Da Pieve C (2005) Dalton Trans 1637–1643 3. Geierstanger BH, Kagawa TF, Chen S-L, Quigley GJ, Ho PS
(1991) J Biol Chem 266:20185–20191 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 νRF-ν31P/ MHz D E F
Fig. 6 X-band Mims ENDOR spectra of Cu(II)–5¢-GMP (sample 2) taken at observer positions D–F (see Fig.2, spectra b). Thin traces experiments, thick traces simulations
Fig. 7 Molecular model of the Cu(H2O)5–5¢-GMP complex consis-tent with ENDOR and HYSCORE data. The copper ion is shown in green and the phosphorus ion in yellow. The figure was created with the help of the program Hyperchem 7.5.1 (Hypercube, Gainesville, FL, USA)
4. Hiai S, Suzuki K, Moriguchi E (1965) J Mol Biol 11:672–691 5. Sundaralingam M, Carrabine JA (1971) J Mol Biol 61:287–309 6. Farrar JA, Neese F, Lappalainen P, Kroneck PMH, Saraste M,
Zumft WG, Thomson AJ (1996) J Am Chem Soc 118:11501– 11514
7. Andersson KK, Schmidt PP, Katterle B, Strand KR, Palmer AE, Lee SK, Solomon EI, Graslund A, Barra AL (2003) J Biol Inorg Chem 8:235–247
8. Calle C, Sreekanth A, Fedin MV, Forrer J, Garcia-Rubio I, Gromov IA, Hinderberger D, Kasumaj B, Leger P, Mancosu B, Mitrikas G, Santangelo MG, Stoll S, Schweiger A, Tschaggelar R, Harmer J (2006) Helv Chim Acta 89:2495–2521
9. delaFuente M, Cozar O, David L, Navarro R, Hernanz A, Bratu I (1997) Spectrochim Acta Part A 53:637–641
10. Weckhuysen BM, Leeman H, Schoonheydt RA (1999) Phys Chem Chem Phys 1:2875–2880
11. Kaczmarek P, Jezowska-Bojczuk M (2005) Inorg Chim Acta 358:2073–2076
12. Hoogstraten CG, Grant CV, Horton TE, DeRose VJ, Britt RD (2002) J Am Chem Soc 124:834–842
13. Schweiger A, Jeschke G (2001) Principles of pulse electron paramagnetic resonance. Oxford University Press, Oxford 14. Deligiannakis Y, Louloudi M, Hadjiliadis N (2000) Coord Chem
Rev 204:1–112
15. Davies ER (1974) Phys Lett A 47:1–2
16. Mims WB (1965) Proc R Soc Lond 283:452–457 17. Stoll S, Schweiger A (2006) J Magn Reson 178:42–55
18. Madi ZL, Van Doorslaer S, Schweiger A (2002) J Magn Reson 154:181–191
19. Dikanov SA, Tsvetkov YD, Bowman MK, Astashkin AV (1982) Chem Phys Lett 90:149–153
20. Sigel H, Massoud SS, Corfu NA (1994) J Am Chem Soc 116:2958–2971
21. Sigel H, Song B (1996) Met Ions Biol Syst 32:135–205 22. McGarvey BR (1966) In: Carlin RL (ed) Transition metal
chemistry. Dekker, New York, pp 89–201
23. Peisach J, Blumberg WE (1974) Arch Biochem Biophys 165:691–708
24. Schosseler PM, Wehrli B, Schweiger A (1997) Inorg Chem 36:4490–4499
25. Atherton NM, Horsewill AJ (1979) Mol Phys 37:1349–1361 26. Manikandan P, Epel B, Goldfarb D (2001) Inorg Chem 40:781–
787
27. Dikanov SA, Xun LY, Karpiel AB, Tyryshkin AM, Bowman MK (1996) J Am Chem Soc 118:8408–8416
28. Garcia MLS, Smith JAS (1983) J Chem Soc Perkin Trans II 1401–1408
29. Stoll S, Calle C, Mitrikas G, Schweiger A (2005) J Magn Reson 177:93–101
30. Zanker PP, Jeschke G, Goldfarb D (2005) J Chem Phys 122:024515
31. Sletten E, Lie B (1976) Acta Crystallogr Sect B 32:3301–3304 32. Allen FH, Kennard O (1993) Chem Des Automat News 8:31–37 33. Sheldrick WS (1981) Acta Crystallogr Sect B 37:1820–1824
